McGill.CA / Science / Department of Physics

CPM Seminar

First-Principles Approach for Investigations of Structural and Electronic Properties and Energy Level Alignment at Aqueous Semiconductor Interfaces

James Muckerman

Chemistry Department
Brookhaven National Laboratory

Water splitting using semiconductor-based heterogeneous photocatalysis plays a key role in a promising path to clean and sustainable energy production. Domen's group has reported that the band gap of GaN can be reduced to absorb visible instead of UV light by alloying GaN with ZnO. The band-gap-narrowed GaN/ZnO alloy is an efficient visible-light photocatalyst, although microscopic models for reaction sites and mechanisms remain as important open questions. Using density-functional-theory-based molecular dynamics, we investigate the microscopic structural and electronic properties of aqueous interfaces of nonpolar Wurtzite facets of GaN, ZnO, and representative GaN/ZnO alloy structures. We find that water adsorption is substantially dissociative. At the equilibrated interfaces, most of the surface anions are protonated, while many surface cations are bonded to hydroxide ions. Surface N-sites show stronger basic character and are protonated more readily than surface O-sites. All surface Ga atoms are bonded to hydroxide ions while about 50% of surface Zn atoms are bonded to hydroxide ions. Our earlier work suggests that water oxidation at the GaN-water interface is driven by the localization of photogenerated holes on the adsorbed hydroxides. Additionally, the hard-wall interface presented by the semiconductor imparts ripples in the density of the water. Beyond a 3 Å distance from the semiconductor surface, the water exhibits a bulk-like hydrogen bond network and oxygen-oxygen radial distribution function. Taken together, these characteristics represent the resting (or “dark”) state of the catalytic interface.

GaN, ZnO, and 2 alloys Another important issue is that the relative alignment of the semiconductor band edge and the corresponding redox level in the solvent for a target reaction determines thermodynamically whether photoexcited carriers in the semiconductor can drive the reaction and with what range of overpotential. This influences the design of electrochemical devices for solar energy harvesting. In particular, it is an unavoidable constraint in the search for materials that can serve both as efficient absorbers of the solar spectrum and to supply electrons and holes with sufficient energy to drive relevant reactions, e.g., the hydrogen evolution reaction or the water oxidation reaction. I will present the results of our recent studies of these issues, as well as results from complementary models of the water oxidation reaction at a GaN photoanode.

Thursday, December 11th 2014, 15:30
Ernest Rutherford Physics Building, R.E. Bell Conference Room (room 103)